|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 100, |
|
"global_step": 85, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.058823529411764705, |
|
"grad_norm": 6.165916919708252, |
|
"learning_rate": 1.1111111111111113e-05, |
|
"loss": 2.8208, |
|
"mean_token_accuracy": 0.472363967448473, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.11764705882352941, |
|
"grad_norm": 2.7514894008636475, |
|
"learning_rate": 1.999145758387301e-05, |
|
"loss": 2.2167, |
|
"mean_token_accuracy": 0.5186272092163563, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.17647058823529413, |
|
"grad_norm": 3.146514415740967, |
|
"learning_rate": 1.9694002659393306e-05, |
|
"loss": 1.711, |
|
"mean_token_accuracy": 0.5845877155661583, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.23529411764705882, |
|
"grad_norm": 1.0160449743270874, |
|
"learning_rate": 1.898390981891979e-05, |
|
"loss": 1.3648, |
|
"mean_token_accuracy": 0.644020289182663, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.29411764705882354, |
|
"grad_norm": 0.6413612365722656, |
|
"learning_rate": 1.789140509396394e-05, |
|
"loss": 1.1651, |
|
"mean_token_accuracy": 0.683531029522419, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.35294117647058826, |
|
"grad_norm": 0.5301786661148071, |
|
"learning_rate": 1.646299237860941e-05, |
|
"loss": 1.0656, |
|
"mean_token_accuracy": 0.7045414000749588, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.4117647058823529, |
|
"grad_norm": 0.4749670922756195, |
|
"learning_rate": 1.4759473930370738e-05, |
|
"loss": 1.0082, |
|
"mean_token_accuracy": 0.7174434214830399, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.47058823529411764, |
|
"grad_norm": 6.938104629516602, |
|
"learning_rate": 1.2853362242491054e-05, |
|
"loss": 0.9774, |
|
"mean_token_accuracy": 0.7241898372769355, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.5294117647058824, |
|
"grad_norm": 8.241913795471191, |
|
"learning_rate": 1.0825793454723325e-05, |
|
"loss": 0.9713, |
|
"mean_token_accuracy": 0.7244970127940178, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.5882352941176471, |
|
"grad_norm": 1.2930790185928345, |
|
"learning_rate": 8.763073687306523e-06, |
|
"loss": 0.9613, |
|
"mean_token_accuracy": 0.7273636788129807, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.6470588235294118, |
|
"grad_norm": 3.2476096153259277, |
|
"learning_rate": 6.7530053079531664e-06, |
|
"loss": 0.9469, |
|
"mean_token_accuracy": 0.7303484156727791, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.7058823529411765, |
|
"grad_norm": 2.7763962745666504, |
|
"learning_rate": 4.881149509103993e-06, |
|
"loss": 0.941, |
|
"mean_token_accuracy": 0.7318810477852822, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.7647058823529411, |
|
"grad_norm": 7.312466621398926, |
|
"learning_rate": 3.2271842837425917e-06, |
|
"loss": 0.9448, |
|
"mean_token_accuracy": 0.7313860774040222, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.8235294117647058, |
|
"grad_norm": 7.457351207733154, |
|
"learning_rate": 1.861512827298051e-06, |
|
"loss": 0.9272, |
|
"mean_token_accuracy": 0.7352052882313729, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.8823529411764706, |
|
"grad_norm": 2.955688238143921, |
|
"learning_rate": 8.42266733449425e-07, |
|
"loss": 0.9322, |
|
"mean_token_accuracy": 0.7336135596036911, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.9411764705882353, |
|
"grad_norm": 1.3166555166244507, |
|
"learning_rate": 2.1283154672645522e-07, |
|
"loss": 0.9214, |
|
"mean_token_accuracy": 0.7363106966018677, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 2.04097056388855, |
|
"learning_rate": 0.0, |
|
"loss": 0.9302, |
|
"mean_token_accuracy": 0.7341954424977303, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 85, |
|
"total_flos": 71084393103360.0, |
|
"train_loss": 1.2238820973564597, |
|
"train_runtime": 7471.3701, |
|
"train_samples_per_second": 1.456, |
|
"train_steps_per_second": 0.011 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 85, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": false, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 71084393103360.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|